16:00:02 <smcginnis> #startmeeting Cinder
16:00:04 <openstack> Meeting started Wed Oct  5 16:00:02 2016 UTC and is due to finish in 60 minutes.  The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:06 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:08 <openstack> The meeting name has been set to 'cinder'
16:00:11 <Swanson> Hi
16:00:11 <smcginnis> Hello all
16:00:11 <thingee> o/
16:00:15 <smcginnis> https://wiki.openstack.org/wiki/CinderMeetings
16:00:16 <xyang2> hi
16:00:17 <jseiler> hi
16:00:18 <jgregor> Hello!
16:00:22 <smcginnis> dulek duncant eharney geguileo winston-d e0ne jungleboyj jgriffith thingee smcginnis hemna xyang1 tbarron scottda erlon rhedlind jbernard _alastor_ bluex patrickeast dongwenjuan JaniceLee cFouts Thelo vivekd adrianofr mtanino yuriy_n17 karlamrhein diablo_rojo jay.xu jgregor baumann rajinir wilson-l reduxio wanghao thrawn01 chris_morrell stevemar watanabe.isao,tommylike.hu mdovgal
16:00:22 <slayer> Hello
16:00:28 <dulek> o/
16:00:30 <mtanino> hi
16:00:33 <e0ne> hi
16:00:39 <scottda> hey
16:00:59 <cFouts> hi
16:01:00 <jgriffith> helllloooooooooo
16:01:05 <bswartz> .o/
16:01:08 <geguileo> o/
16:01:08 <patrickeast> Hey
16:01:32 <smcginnis> #topic Summit Session Planning
16:01:45 <smcginnis> #link https://etherpad.openstack.org/p/ocata-cinder-designsummit-planning Planning etherpad
16:01:54 <tbarron> hi
16:02:01 <smcginnis> We need to finalize our sessions for the summit.
16:02:18 <smcginnis> We have more slots than proposed topics at the moment.
16:02:29 <smcginnis> If we don't need all of those, I'd like to give those back ASAP.
16:02:41 <thingee> smcginnis: we wouldn't mind taking them back if needed :)
16:02:41 <smcginnis> I think there were a couple other teams looking for space.
16:02:46 <smcginnis> thingee: ;)
16:02:53 <smcginnis> thingee: I'm sure that would make a few people happy.
16:02:59 <e0ne> jgriffith: do you want to continue a discussing about standalone cinder (SDS)?
16:03:08 <thingee> less meeting time is cool with me. More time to enjoy spain
16:03:15 <flip214> e0ne: +1
16:03:15 <smcginnis> There you go!
16:03:18 <mtanino> +1
16:03:26 <jgriffith> e0ne: sure
16:03:26 <hemna> doink
16:03:37 <smcginnis> jgriffith: Should I put you down on there to lead that?
16:03:37 <e0ne> ok. adding it to the etherpad
16:03:46 <jungleboyj> e0ne: I would be interested in that ... don't ask.
16:03:55 <smcginnis> jungleboyj: :)
16:04:04 <jungleboyj> smcginnis: You have no idea.
16:04:08 <_alastor_> o/
16:04:16 <dulek> Do we have any outstanding stuff from the mid-cycle?
16:05:02 <e0ne> I would like to propose backup discussion again
16:05:16 <hemna> dulek, we still don't have a way to fix bugs in drivers for unsupported OS releases.
16:05:19 <smcginnis> e0ne: Great. Want to add that to the etherpad as well?
16:05:22 * diablo_rojo sneaks in the back of the room
16:05:26 <smcginnis> hemna: Got that on there.
16:05:28 <hemna> I think this is terrible.
16:05:35 <e0ne> I've got few specs under the review and we need to figure out what state of backup scaling is
16:05:41 <smcginnis> hemna: I'm going to send a ML announcement to try to get some vendors to that one.
16:05:42 <e0ne> smcginnis: in progress
16:05:48 <hemna> smcginnis, ok thanks.
16:05:59 <hemna> I'm super bummed I won't be there to advocate for this stuff.
16:05:59 <smcginnis> Would like to get the input from those actually providing support for older releases.
16:06:11 <smcginnis> hemna: Oh right, I keep forgetting that.
16:06:46 <smcginnis> We also have the afternoon working session to go over smaller topics that inevitably come up.
16:06:47 <hemna> I could just mail someone a chewbacca stuffed animal and put him on the table in my place.
16:06:53 <smcginnis> ;)
16:07:00 <dulek> I wonder if a fishbowl on gathering feedback about Cinder future wouldn't be beneficial.
16:07:05 <smcginnis> I should find out...
16:07:09 <jungleboyj> :-)  We will keep a chair by the camera for you hemna
16:07:16 <hemna> jungleboyj, thanks
16:07:17 <dulek> We're doing that at the mid-cycles, but there's no operators mostly.
16:07:30 <smcginnis> #startvote Will you be able to attend the Friday afternoon session? Yes, No
16:07:31 <openstack> Begin voting on: Will you be able to attend the Friday afternoon session? Valid vote options are Yes, No.
16:07:32 <openstack> Vote using '#vote OPTION'. Only your last vote counts.
16:07:32 <dulek> s/there's/there are
16:07:41 <smcginnis> Informal poll to get an idea..
16:07:45 <scottda> We actually need to make sure we have a camera, since we usually count on hemna
16:07:48 <e0ne> yes
16:08:00 <dulek> #vote Yes
16:08:02 <scottda> #vote Yes
16:08:03 <patrickeast> #vote Yes
16:08:05 <erlon> hey
16:08:05 <smcginnis> Please use #vote with Yes or No so we can get an idea of how many people will be able to stick around Friday afternoon.
16:08:05 <xyang2> #vote Yes
16:08:09 <diablo_rojo> #vote yes
16:08:09 <e0ne> #vote Yes
16:08:09 <smcginnis> #vote Yes
16:08:19 <bswartz> #vote no
16:08:23 <jungleboyj> #vote yes
16:08:25 <jgriffith> #vote yes
16:08:29 <diablo_rojo> I will have a camera and the tripod
16:08:33 <smcginnis> Looks good so far. I was afraid too many people had already booked their return flight for Friday evening.
16:08:33 <diablo_rojo> scottda, ^
16:08:41 <scottda> diablo_rojo: Cool. Thanks.
16:08:53 <diablo_rojo> Don't have a fancy speaker/microphone though
16:09:07 <jungleboyj> diablo_rojo: I think Sean has one.
16:09:07 <smcginnis> diablo_rojo: I might have one that kind of works.
16:09:10 <diablo_rojo> hemna, can you get me access to the youtube channel so I can stream?
16:09:16 <_alastor_> #vote yes
16:09:24 <smcginnis> diablo_rojo: I can probably add you to it.
16:09:32 <erlon> #vote  Yes
16:09:40 <hemna> I think smcginnis can do it
16:09:43 <smcginnis> OK, anybody else going to the summit that needs to vote yet?
16:10:06 <smcginnis> #endvote
16:10:07 <openstack> Voted on "Will you be able to attend the Friday afternoon session?" Results are
16:10:08 <openstack> Yes (11): scottda, smcginnis, erlon, jungleboyj, jgriffith, _alastor_, diablo_rojo, e0ne, xyang2, patrickeast, dulek
16:10:09 <openstack> No (1): bswartz
16:10:24 <smcginnis> bswartz: Fine, be that way. :P
16:10:35 <smcginnis> Cool, better turn out than I was afraid we would have.
16:10:46 <smcginnis> OK, back to the actual session scheduling...
16:10:47 <erlon> smcginnis: just confirming friday afternoon will be the happyhour isnt?
16:10:53 <erlon> :)
16:10:57 <smcginnis> :)
16:11:07 <bswartz> lol
16:11:12 <smcginnis> We need to schedule an informal cinder team gather too.
16:11:16 <smcginnis> But anyway..
16:11:18 <dulek> Hey, will there be any evening event besides marketplace mixer? ;>
16:11:20 <bswartz> it's just that all the meetups are in the same timeslot and I kinda have to do the manila one
16:11:27 <smcginnis> Back to here: https://etherpad.openstack.org/p/ocata-cinder-designsummit-planning
16:11:42 <smcginnis> bswartz: Eh, they can handle it without you. :D
16:11:51 <thingee> informal team gather, just pick any random tapas place. can't go wrong
16:11:51 <bswartz> I still love you guys
16:11:57 <e0ne> dulek: I believe we'll have cinder&beer mettup, smcginnis won't we?
16:12:12 <smcginnis> e0ne: Definitely. We should do Tuesday or Wednesday night.
16:12:36 <bswartz> thingee: make sure that people know that tapas in spain are NOT the same as tapas in the USA
16:12:55 <smcginnis> Please take a look at the proposed schedule.
16:13:04 <smcginnis> This is just a strawman.
16:13:14 * erlon goes to check the urban dictionary
16:13:52 <e0ne> smcginnis: not sure that ' Cinder backup improvements' is OK  for fishbowl, we can discuss it in a smaller room too
16:14:08 <smcginnis> The last scheduled slot on Friday I put to talk about priorities and schedule. We could move that into working group time.
16:14:38 <smcginnis> e0ne: Yeah, wasn't too sure about that. We could move that into that last Friday slot, then free up a fishbowl for either some other bigger topic or give it back.
16:14:51 <e0ne> smcginnis: +1
16:15:12 <dulek> Cinder-Nova API shouldn't be a fishbowl?
16:15:22 <dulek> So we can fit whole Nova team in the room.
16:15:49 <smcginnis> dulek: That one I think will be our internal discussion of John's patches and what we see from the Cinder perspective.
16:16:07 <dulek> Oh, okay, fine with me then.
16:16:10 <smcginnis> dulek: Then on Friday morning Nova is going to host the actual Cinder and Nova joint session.
16:16:21 <smcginnis> Hopefully that is a big room.
16:16:29 <ildikov> smcginnis: +1
16:16:36 <thingee> bswartz: ya baby eels :)
16:16:51 <ildikov> smcginnis: we would need a bigger room for that
16:16:59 <smcginnis> Hah, out of context that's a weird statement thingee ;)
16:17:13 <thingee> smcginnis: re: tapas in spain are not the same as the US
16:17:27 <smcginnis> thingee: Yeah, got it. Still funny. :]
16:18:02 <smcginnis> Any other thoughts for topics? Or should we give back a fishbowl?
16:18:42 <smcginnis> Since O is short and mostly bugfix we should probably spend a little time thinking ahead to Pike. But not sure if that is worth a fishbowl.
16:19:12 <jungleboyj> Pike will be our opportunity to Peak!
16:19:15 <dulek> We're speaking about future at the mid-cycles. Maybe such discussion can benefit from broader fishbowl session?
16:19:27 <scottda> dulek: +1
16:19:30 <dulek> I mean - mid-cycles gather mostly developers.
16:19:39 <dulek> Summit has broader attention.
16:19:48 <erlon> will that still exist?
16:19:57 <smcginnis> erlon: The PTG will.
16:20:18 * jungleboyj is going to miss our mid-cycles!
16:20:25 * diablo_rojo agrees
16:21:00 <dulek> jungleboyj: Hey, colocated mid-cycle in Sydney can't turn out bad. ;)
16:21:00 <smcginnis> So... am I hearing we want a fishbowl to be able to gather more than just our developer input on Pike feature planning?
16:21:05 <scottda> might want a mid-cycle, with PTGs in Feb and Sept
16:21:10 <erlon> may be the PTG will be a BIG mid-cicle
16:21:29 <dulek> erlon: PTG will be more like BIG design summit
16:21:35 <scottda> But PTG won't be mid-cycle, it's start-of-cycle
16:21:53 <smcginnis> In case folks aren't aware: http://www.openstack.org/ptg
16:22:04 <erlon> scottda: hmmm, and the summit will be in the middle of the release?
16:22:16 <bswartz> yeah the timing of the midcycle will be close to or overlapping the actual conferences
16:22:28 <scottda> Pike PTG => Q PTG is 8 months apart
16:22:29 <dulek> smcginnis: Maybe we should vote on last fishbowl destiny? :)
16:22:31 <smcginnis> Good diagram if you follow that link.
16:22:34 <bswartz> the PTG will be start/end of cycle
16:22:36 * dulek tries to get responsibility off him.
16:22:45 <smcginnis> dulek: ;)
16:23:21 <bswartz> scottda: I count 6.5 months not 8
16:23:55 <hemna> are folks even going to go to the summits ?
16:23:58 <scottda> bswartz: OH, you're right
16:24:10 <Swanson> math
16:24:19 <xyang2> smcginnis: maybe we can have a mini mid-cycle at the Pike summit.  don't we have an extra day now without design sessions?
16:24:22 <jgriffith> hemna: remains to be seen
16:24:28 <smcginnis> #startvote What should we do with our last fishbowl? Planning, Return, Horde
16:24:29 <openstack> Begin voting on: What should we do with our last fishbowl? Valid vote options are Planning, Return, Horde.
16:24:29 <hemna> maybe we should do midcycles during the summits, since they don't seem to be of much use to us anymore
16:24:30 <openstack> Vote using '#vote OPTION'. Only your last vote counts.
16:24:31 <smcginnis> xyang2: Maybe
16:24:38 <jgriffith> hemna: I'll more than likely be replaced by a marketing expert
16:24:46 <bswartz> hemna: summit won't exist
16:24:50 <dulek> Wait, what do the options mean?
16:24:51 <diablo_rojo> hemna, I was thinking the same thing
16:24:59 <jgriffith> bswartz: ok... "conference"
16:24:59 <erlon> hemna: +1
16:25:12 <smcginnis> Planning - use it to plan Pike stuff. Return - give it back for someone else to use.
16:25:13 <jgriffith> bswartz: hemna at least I'm assuming that's what hemna menat
16:25:15 <jgriffith> meant
16:25:22 <hemna> jgriffith, that's what I was worried about and brought up to johnathan in Austin.  not many people cared.
16:25:22 <e0ne> hemna: good idea, but could be hard to get budget approved for it
16:25:23 <bswartz> oh wait they are calling the conference a "summit" on that page -- how confusing
16:25:29 <smcginnis> Horde - be selfish and hold on to it assuming we'll have something important come up.
16:25:39 <smcginnis> Mostly joking about the Horde option.
16:25:43 <diablo_rojo> hemna, We spoke up too late
16:25:46 <erlon> jgriffith: if you will need to go to the MC somewhere else is the same to go along with the summit
16:25:48 <jungleboyj> #vote Planning
16:25:50 <hemna> he looked at me like I was an alien, when I brought it up
16:25:52 <xyang2> #vote Planning
16:25:56 <bswartz> s/horde/hoard/
16:25:57 <scottda> #vote Planning
16:26:02 <dulek> #vote Planning
16:26:05 <e0ne> #vote Planning
16:26:24 <smcginnis> bswartz: :)
16:26:41 <smcginnis> #vote Planning
16:26:44 <jgriffith> #vote Return
16:26:45 <hemna> #vote drink
16:26:45 <_alastor_> #vote Planning
16:26:45 <openstack> hemna: drink is not a valid option. Valid options are Planning, Return, Horde.
16:26:49 <hemna> damn
16:26:54 <e0ne> :(
16:26:57 <smcginnis> !!
16:26:58 <openstack> smcginnis: Error: "!" is not a valid command.
16:27:03 <hemna> sheeeeit
16:27:04 <scottda> That bot is a real kill-joy
16:27:04 <smcginnis> Darn, still no table flip.
16:27:05 <e0ne> !
16:27:11 <bswartz> cinder fishbowls are always "hordes" based on past experience
16:27:13 <diablo_rojo> #vote planning
16:27:24 <hemna> (╯°□°)╯︵ ┻━┻
16:27:30 <smcginnis> hemnabot
16:27:34 <erlon> hemna: haha I love that!
16:27:41 <jgriffith> LOL
16:27:45 <diablo_rojo> The bot is a teetotaler
16:27:50 <hemna> :)
16:27:51 <erlon> #vote planning
16:28:14 <smcginnis> Anyone need a little more time?
16:28:38 <smcginnis> #endvote
16:28:39 <openstack> Voted on "What should we do with our last fishbowl?" Results are
16:28:40 <openstack> Planning (9): scottda, smcginnis, erlon, jungleboyj, _alastor_, diablo_rojo, e0ne, xyang2, dulek
16:28:41 <openstack> Return (1): jgriffith
16:29:04 <jgriffith> (╯°□°)╯︵ ┻━┻
16:29:04 <jgriffith> [10:27:29]
16:29:09 <hemabot> (╯°□°)╯︵ ┻━┻
16:29:11 <hemabot> [10:27:29](╯°□°)╯︵ ┻━┻
16:29:19 <smcginnis> :)
16:29:21 <hemna> :)
16:29:28 <e0ne> smcginnis: let's move planning session to the last of fishbowl's
16:29:30 <jgriffith> haha.. didnt' go as smoothly as I'd planned
16:29:34 <smcginnis> OK, we'll use it for planning.
16:30:06 <hemabot> [10:27:29]/join #openstack-cinder
16:30:11 <smcginnis> e0ne: OK, and have the stand alone cinder discussion right before it I think.
16:30:20 <e0ne> smcginnis: LGTM
16:30:46 <jungleboyj> (╯°□°)╯︵ ┻━┻
16:31:00 <diablo_rojo> This has to be one of the best meeting logs ever
16:31:12 <smcginnis> ┬─┬ノ( º _ ºノ)
16:31:17 <smcginnis> Let's settle down now. :)
16:31:23 <thingee> smcginnis: ha was just about to do that
16:31:24 <jungleboyj> There are so many good ones for Cinder!
16:31:28 <jungleboyj> smcginnis: No!  (╯°□°)╯︵ ┻━┻
16:31:33 <smcginnis> OK, please take a look at the proposed schedule.
16:31:34 <smcginnis> thingee: ;)
16:31:41 <thingee> ┻━┻ ︵ヽ(`Д´)ノ︵ ┻━┻
16:31:47 <smcginnis> If this looks OK, I will update the session info.
16:31:52 <smcginnis> thingee: Oooo, nice!
16:32:11 <thingee> so many tables
16:32:11 <jungleboyj> thingee: That is awesome!
16:32:15 <diablo_rojo> thingee, got one for flipping a laptop? ;)
16:32:29 <jungleboyj> diablo_rojo: Oooh ...
16:32:55 <thingee> still think lenovo should cover it for anyone working in open source
16:33:05 <jgriffith> smcginnis: if we happen to save 10 minutes to talk about Attach changes that would be awesome
16:33:17 <jgriffith> smcginnis: if we can stop flipping tables that is :)
16:33:23 <diablo_rojo> thingee at least it makes a good story
16:33:30 <smcginnis> jgriffith: I was planning on the Thursday 9:50 session being all about that.
16:33:35 <jungleboyj> diablo_rojo: ++
16:33:42 <jgriffith> smcginnis: haha.. I meant here today :)
16:33:53 <smcginnis> jgriffith: Oh, absolutely.
16:34:00 <diablo_rojo> thingee, Will you have one last round of contraband cinder logo stickers?
16:34:04 <smcginnis> #topic Attach/detach API changes
16:34:05 <ildikov> jgriffith: +1 :)
16:34:17 <smcginnis> jgriffith: Want to give an update?
16:34:25 <jgriffith> so I'll try and make this short and open up a discussion
16:34:38 <jgriffith> I've been on and off working on our attach/detach flow
16:35:02 <hemna> :) you can blame me.
16:35:07 <thingee> diablo_rojo: nope
16:35:12 <jgriffith> I have a version that basically just condenses initialize/attach and terminate/detach into two command:  create_attachment and remove_attachment
16:35:19 <thingee> diablo_rojo: I mean, I guess I could print 'em
16:35:27 <jungleboyj> thingee: Sadness.
16:35:45 <jungleboyj> Need a bunch to hold on to for the future.
16:35:47 <jgriffith> https://review.openstack.org/#/c/327408/
16:35:53 <smcginnis> jgriffith: Do you have links to the two... nevermind.
16:36:05 <thingee> vintage. I don't even have a sticker
16:36:26 <smcginnis> #link https://review.openstack.org/#/c/327408/
16:36:27 <jungleboyj> thingee: Well, there you go,  You need to print more.
16:36:41 <jgriffith> There's also WIP of nova https://review.openstack.org/#/c/330285/
16:36:53 <jgriffith> and cinderclient:  https://review.openstack.org/#/c/327409/
16:37:20 <jgriffith> so this is pretty close to working, some details I need to fix in unshelve and live-migration
16:37:40 <smcginnis> jgriffith: There was another patch for a slightly different approach, right?
16:37:48 <smcginnis> jgriffith: Or should we just focus on this for now?
16:37:50 <jgriffith> Yes...
16:38:15 <jgriffith> So I started looking at something else as well that is more of a *modification* of how cinder works more than antyhing else
16:38:24 <jgriffith> https://review.openstack.org/#/c/382045/
16:38:41 <smcginnis> #link https://review.openstack.org/#/c/382045/
16:38:46 <jgriffith> each approach has pros/cons
16:38:58 <smcginnis> #link  https://review.openstack.org/#/c/330285/ Nova changes
16:39:12 <smcginnis> #link  https://review.openstack.org/#/c/327409/ python-cinderclient changes
16:39:25 <jgriffith> V2 provides a little bit more flexibility and granularity, while also being basically the same flow that we have today
16:39:52 <jgriffith> the only thing it really does is clean up the implementation of the methods, and gets rid of some of the guesswork on parameters etc
16:40:29 <jgriffith> BOTH approaches tie in an attachment object that gets stored and is used as a handle for everything on the Nova side (or any caller)
16:40:59 <jgriffith> the reason for that is to get rid of some of the split brain issues we have currently and make multi-attach simpler
16:41:22 <jgriffith> I'm struggling a bit right now though... trying to determine what the best option might be
16:41:37 <e0ne> /me needs more time to take a look on John's patches:(
16:41:46 <jgriffith> Mostly between the condensed flow and the reworked impl
16:42:20 <jgriffith> What's interesting about the V2 approach... is that if you break it down and compare it to what we have:  http://paste.openstack.org/show/584483/
16:42:25 <ildikov> and it's not only for multi-attach, but there are workarounds in live mogration and shelve offload in Nova that would be great to remove and use something more explicit
16:42:50 <jgriffith> ildikov: right, sorry... I've just been focusing mostly on keeping multi-attach in mind for this
16:43:05 <smcginnis> jgriffith: the Nova team thought the reworked flow might have a benefit, right?
16:43:05 <jgriffith> ildikov: but yes, it does actually simplify  a number of other things
16:43:20 <ildikov> jgriffith: big thanks for that!
16:43:40 <jgriffith> smcginnis: I honestly can't say that I've had a strong response on that; but I can find out
16:44:08 <jgriffith> smcginnis: Also, I haven't floated a nova version of the V2 stuff up to Nova yet to get feedback there either
16:44:21 <ildikov> jgriffith: I just thought to mention that we chose this path for a more generic and hopefully simpler API interaction :)
16:44:38 <jgriffith> ildikov: yes, and that's an excellent point
16:44:54 <jgriffith> ildikov: also reminds me of another advantage of the first proposal :)
16:45:17 <jgriffith> ildikov: V1 is significantly more work up front, V2 is more of a baby step
16:45:35 <smcginnis> I would be for the simpler interaction. Unless there is something critical needed for v2.
16:45:49 <jgriffith> ildikov: V1 IMHO definitely leads to more simplicity in the long run
16:46:10 <smcginnis> jgriffith: So maybe we should keep v2 in our back pocket and just move forward pushing v1 as our preferred solution to this.
16:46:14 <jgriffith> smcginnis: ildikov the only thing that came up was a conern by johnthetubaguy about shared connections
16:46:15 <patrickeast> jgriffith: i was a little concerned about that part, i was looking through the latest ones last night and was wondering how long the v2 one would stick around before we want to do something more simple and go the route of v1 anyway
16:46:17 <hemna> jgriffith, I think since v1 seems to be a simpler approach in the long run, I'd say +1 to it.
16:46:28 <patrickeast> which means supporting 3 different attach api's forever. .woo!
16:46:37 <jgriffith> patrickeast: yeah, that's fair and I def agree
16:46:55 <smcginnis> Right, we should get this right (as much as possible) and hopefully be done with it.
16:46:56 <dulek> patrickeast: Forever…?
16:46:58 <ildikov> jgriffith: I would go for simplicity if we can, but it's a bit of a balancing act here that I can understand
16:47:03 <scottda> Yes, and v1 uses entirely new methods, which means less confusion to support and understand the changes.
16:47:11 <smcginnis> scottda: +1
16:47:16 <patrickeast> dulek: mostly joking ;)
16:47:16 <hemna> dulek, once an API is in place, it's practically in place forever.
16:47:38 <smcginnis> scottda: That's actually a very good point. That would be a clean break and avoid some confusion.
16:47:38 <ildikov> scottda: +1
16:47:41 <jgriffith> ok... I think most of the on the fly input here is confirming some things for me so that's great
16:47:52 <dulek> hemna: But this one is rather internal, so I think it's possible to find some exception to "forever".
16:47:56 <jgriffith> I'll continue on the V1 path today
16:47:58 <dulek> hemna: Almost forever maybe? ;)
16:48:04 <hemna> jgriffith, +1
16:48:11 <ildikov> jgriffith: +1
16:48:17 <jgriffith> dulek: patrickeast hemna FWIW note in my V1 I'm deprecating the methods
16:48:18 <scottda> jgriffith: Thanks for the work on this.
16:48:33 <jgriffith> dulek: I plan to remove them once there's no n-1 delta between cinder/nova
16:48:47 <smcginnis> We only have to support Nove-2, so we would be able to get rid of it eventually.
16:48:51 <dulek> jgriffith: This may be a little too soon.
16:48:54 <jgriffith> I also plan to rewrite the initialize/attach and terminate/detach in V1 as follow up
16:48:57 <smcginnis> jgriffith: Oh, is it n-1 or n-2?
16:49:13 <jgriffith> dulek: right, I'm not removing them... just logging a warning that "it's going to happen some day"
16:49:13 <ildikov> can we say that there's a team agreement on V1 and document it or let's wait with decisions?
16:49:20 <dulek> jgriffith: Suuure. :)
16:49:39 <dulek> Just wanted to say that some people run very different combinations of Nova and Cinder versions.
16:49:41 <jgriffith> smcginnis: I thought it was n-1, but either way it'll work, you just delay longer :)
16:49:48 <jgriffith> dulek: yeah, that's fair
16:49:50 <smcginnis> ildikov: Let's just proceed with v1 until we find a reason that we have to move to v2.
16:49:56 <smcginnis> jgriffith: yep
16:50:05 <jgriffith> smcginnis: ok, sounds great
16:50:08 <jgriffith> Thanks everyone!!
16:50:14 <ildikov> smcginnis: fair enough
16:50:16 <jgriffith> I'll get some updates pushed today
16:50:24 <smcginnis> Dang, one item on the agenda and somehow we're 50 minutes in. :)
16:50:27 <ildikov> jgriffith: thanks for the work on this!
16:50:30 <smcginnis> jgriffith: Thanks for working on all of that.
16:50:40 <jgriffith> yeah, np
16:50:55 <smcginnis> #topic Open discussion
16:51:05 <smcginnis> Anything else to discuss?
16:51:09 <smcginnis> Oh, I have one thing.
16:51:20 <smcginnis> I know quite a few people are travelling early for the summit.
16:51:23 <jungleboyj> A rhetorical question.
16:51:42 <smcginnis> Should we skip the meeting next week, or would it be good to still have it for last minute prep?
16:52:08 <scottda> There's 2 meetings before the summit, right?
16:52:30 <smcginnis> scottda: Oh right, I'm getting ahead of myself.
16:52:35 <smcginnis> So next week is definitely on.
16:52:40 * jungleboyj was wondering
16:52:48 <smcginnis> But what about the week before?
16:53:06 <smcginnis> Maybe we can discuss next week once we're closer and have a better idea if we need to meet or not.
16:53:10 <Swanson> that would be this week
16:53:11 <jungleboyj> So, next week yes, probably not the following week.
16:53:19 <erlon> smcginnis: may be if somone adds something to the agenda?
16:53:44 <erlon> smcginnis: otherwise we skip it
16:53:46 <smcginnis> erlon: Sure. We can wait and see how it goes. If nothing proposed to discuss I will send out a notice to cancel.
16:53:59 <smcginnis> OK, anything else for this week?
16:54:42 <jungleboyj> No!  (╯°□°)╯︵ ┻━┻
16:55:00 <smcginnis> Alrighty then. Thanks everyone!
16:55:00 <erlon> hahaha
16:55:14 * jungleboyj could resist getting the last table in.
16:55:17 <smcginnis> #endmeeting